Charting the Course Ahead: The Future of AI in Marine Cargo Survey Businesses

The marine industry has long relied on the expertise and knowledge of cargo surveyors to ensure the safe and efficient transport of goods across the world’s oceans. In today’s rapidly evolving information age, knowledge and expertise is not just power—it is a strategic asset that drives innovation, growth, and competitive advantage. As marine survey businesses navigate an increasingly complex landscape, the integration of artificial intelligence (AI) has the potential to revolutionise the way knowledge is harnessed, analysed and applied across all industries, and the marine survey industry is no different.

The foundation of knowledge businesses – such as marine surveying- lies in the ability to efficiently access, organise, and interpret vast amounts of data and information and then use that data and information productively and profitably. AI technologies, such as natural language processing, machine learning and semantic analysis are empowering organizations to use data and extract meaningful insights from unstructured data sources such as text documents, research papers and online content.

Moreover, AI-driven knowledge management systems are revolutionising information retrieval and knowledge sharing within organizations. By employing sophisticated algorithms, these systems can automatically tag, categorise and connect relevant pieces of information, allowing for efficient search and retrieval of and then fostering collaboration between teams and facilitating knowledge transfer across organisations.

AI is also propelling knowledge businesses toward data-driven decision-making. By leveraging predictive analytics AI algorithms can identify patterns in data, predict trends and generate accurate forecasts in markets, resourcing needs, skills demands and pricing, empowering businesses to make decisions with confidence. This enables organisations to stay ahead of the competition, anticipate market changes and adapt to dynamic customer demands.

Additionally, AI-powered virtual assistants and chatbots are transforming customer service and support in knowledge businesses. These intelligent agents can provide instant, personalised assistance both inside the business and to its clients. With natural language understanding and machine learning capabilities, virtual assistants offer unparalleled access to support and knowledge and have the potential to transform knowledge businesses in the future.

The forward path of the marine cargo surveying industry will be tied to the development of AI and future applications of AI, there can be no doubt about that. AI’s ability to analyse vast amounts of data, learn from patterns, and support intelligent decision making is poised to revolutionise the way marine businesses work, enhancing efficiency, accuracy, and sustainability.

Before we look into one possible game changing application of AI in marine surveying and consulting, it’s worth defining what we mean by AI, and what we do not currently define as AI.

What Is AI?

The term “artificial intelligence” is often ill-defined and occasionally mis-defined so it’s useful to consider what AI is and it is not, at least at the time of writing this article.

Is AI a computer and algorithm driven system able to perform tasks that would require human intelligence? Is it machines that copy intelligent human behaviour or the simulation of human intelligence processed by machines? Is AI a set of systems that think like humans? Or is it systems that act like humans? Or systems that think and act rationally (in a way that humans may not always act)?

Or is AI a combination of machine learning based on user inputs and deep learning where the machine examines its own algorithms and adjusts them to get better at a task?

AI is all of this, depending on the application and the context of the application. We must be clear about what AI is not at this moment in time and for that we have to split AI into two broad groupings of Weak AI and Strong AI.

Types of AI chart
Types of AI chart

Weak AI – Your Helpful Friend, Today

Weak AI refers to artificial intelligence systems or technologies designed to perform specific tasks or solve specific problems within a limited domain. Unlike strong AI, which aims to mimic human intelligence and possess general intelligence across multiple domains, weak AI focuses on accomplishing well-defined tasks with a narrow scope.

Weak AI systems are built to excel in specific areas and are trained or programmed to perform a particular function or set of functions. They operate based on predefined algorithms and rules and rely on data input to generate outputs or make decisions. These AI systems are task-specific and do not possess the ability to understand or generalise beyond their specific domain.

Examples of weak AI applications are applications that provide speech recognition, image recognition, recommendation systems, virtual assistants, and autonomous vehicles. For instance, voice-activated virtual assistants like Siri or Alexa can understand and respond to user queries within a limited context, but they lack the comprehensive understanding and reasoning capabilities of human intelligence.

Weak AI systems are typically developed using techniques that leverage large datasets to learn patterns and make predictions or classifications within their designated domain of work. However, these systems do not possess consciousness or self-awareness and are not capable of independent thinking or understanding the broader context of their actions.

While weak AI systems may exhibit impressive performance within their specific tasks, they lack the flexibility and adaptability of human intelligence. They are designed to excel in well-defined and constrained environments, making them valuable tools for solving specific problems and enhancing productivity. However, they are limited in their ability to transfer knowledge or generalise their learnings to different domains.

Strong AI

Strong AI, also known as Artificial General Intelligence (AGI) or human-level AI, refers to artificial intelligence systems or technologies that possess the ability to understand, learn, and apply knowledge across a wide range of domains and tasks at a level equal to or exceeding human intelligence. Unlike weak AI, which is designed for specific tasks, strong AI aims to replicate human-like intelligence and cognitive abilities.

The goal of strong AI is to create intelligent machines that exhibit consciousness, self-awareness, and the capacity for independent thought. These systems possess the ability to understand natural language, reason, learn from experience, solve problems, and adapt to new situations. Strong AI seeks to mimic the cognitive processes of human beings and exhibit a level of intelligence that is indistinguishable from human intelligence, capable of autonomous learning and reasoning. They can understand and process information from diverse sources, make sense of complex data, and apply their knowledge to solve problems across various domains.

Artificial Superintelligence (ASI) is an additional strand of Strong AI and refers to a theoretical future state of artificial intelligence where machines surpass human intelligence across virtually all domains and tasks. ASI represents a level of intelligence that exceeds human capabilities to such an extent that it is difficult for humans to comprehend or predict its full range of capabilities and implications.

ASI would possess not only the ability to understand and reason but also the capacity for unlimited learning, creativity, problem-solving, and adaptability. It would be capable of self-improvement, constantly enhancing its own intelligence and surpassing the limitations of its initial programming. ASI would have an unparalleled ability to process and analyse vast amounts of data, make accurate predictions, and generate innovative solutions to complex problems.

The implications of ASI are profound and uncertain. Some argue that ASI could bring unprecedented advancements, leading to solutions for complex global challenges, advancements in science and medicine, and breakthroughs in technological innovation. However, the development of ASI also poses significant risks and challenges. The impact on society, including job displacement, economic disparities and ethical considerations are subjects of intense debate. Safeguarding against the risks associated with ASI requires careful consideration of safety measures, transparency and ethical frameworks.

While still an ongoing pursuit, achieving strong AI would have profound implications requiring careful consideration of its ethical, societal and philosophical impacts, leaving aside the obvious implications for all knowledge industries.

Generative Pretrained Transformers – A Weak AI Making a Difference Now

We’re not living with Strong AI at this point, but Weak AI is everywhere. Almost every consumer smartphone application is a form of Weak AI, supporting human decisions by providing prompts based on the prior behaviour of the user. A simple autocorrect function in a texting application is a form of Weak AI, and it takes little time to identify the other Weak AI that each of uses every day. However, the recent emergence of one particular application is especially worth investigation in the context of the immediate future of marine survey businesses.

One of the most prominent and intriguing emerging Weak AI applications is the Generative Pretrained Transformer, GPT. These have been discussed in the pages of this journal before, but it’s worth digging into GPTs and looking at how they can support businesses now. GPTs are artificial neural networks, pretrained on large data sets of unlabelled text and able to generate novel, human-like content. In effect, the user asks the GPT to perform a task and the GPT produces content the responds to the question that has been asked via a natural language model. Use of a GPT requires absolutely zero technical coding capability and access to a GPT is by a chatbot. Ask it a question – and in particular a detailed and complex technical question – and you’ll get a complex and detailed technical answer.

GPT Cycle
GPT Cycle

GPTs – at present – produce astonishing detail in technical content. At a recent live demonstration of a GPT at the IIMS Conference a GPT produced a detailed procedural plan for the loading of steel coils on to a vessel, and when secondary and tertiary questions were asked the detail of the plan became even more impressive in real time. The potential of that capability should be self evident, but it has to be balanced against the reality that GPTs do not possess understanding or reasoning capabilities and – at this stage of their development – they can act as support to decision making and thereby free up technical expert time to concentrate on the reasoning element of decision making, the procedural piece having been generated by the GPT.

GPTs rely on statistical patterns learned from training data to generate text, and their output is based on the statistical likelihood of certain sequences of words. If under trained they can produce responses that are grammatically corrent but semantically incorrect. However, further pretraining and then specific fine tuning can iron out that problem and there can be little doubt that the GPT application will become a disruptor in the same way that Google began democratising knowledge 25 years ago.

ChatGPT – one of several GPTs – was released in November 2022 with ChatGPT-4 released in March 2023. The latest version of ChatGPT takes images as well as text and produces informational summaries from visuals and diagrams as well as text.

It has been shown to be capable of passing professional examinations at the level of the top 10% of students in fields such as oncology, engineering and plastic surgery. In April 2023 Microsoft and Epic Systems announced plans to provide healthcare providers with GPT-4 powered systems to assist in responses to patient questions and with analysis of patient records. Clearly, something is happening around the latest generation of GPTs and their ability to summarise and contextualise complex technical questions, and everybody in every consulting industry needs to be aware of these tools and their ability to support, impact and disrupt business environments.

Very clearly, the integration of GPTs into consulting type businesses and their use as a procedural conten

t generator at this point is something that all expert professions should take seriously and consider as they build their practices now and into the future.

Five Thoughts on AI In Business

More generally, AI has the potential to transform the marine survey business in numerous ways, offering opportunities for automation, collaboration, augmented decision-making, upskilling, and

changing job roles.

Automation: AI can automate repetitive and mundane tasks, freeing up human resources to focus on more complex and strategic activities. By leveraging machine learning algorithms businesses can streamline repetitive workflows, improve operational efficiency and reduce errors. Automation also minimizes errors and accelerates processes, leading to increased productivity and scal

ability. Obvious applications are the extraction of data from forms, management of invoicing and financial data the production of standard reports and data validation work, all of which take many, many hours of laborious work when done by people. We’d do better freeing people from that kind of drudgery and put them into a position where they can bring their creativity to bear in the business.

Collaboration: AI technologies enable enhanced collaboration among teams and across departments. Intelligent chatbots and virtual assistants are an unbeatable source of rapid information sharing amongst a team and provide a fast and accurate platform for collective problem-solving. Using a GPT in a team setting allows procedural plans to take shape quickly in real time and then frees the team to perform their actual task and get their expertise into play.

Augmented Decision Making: Machine learning algorithms can analyse vast amounts of data, identify patterns, and generate predictions or recommendations. An obvious application would be that of scenario modelling in project cargo operations, where an AI would run several different heavy lift scenarios and simulations to quickly discard expensive or less effective options ahead of time. That frees up hours of project time that can then be spent on more productive tasks, and also points up the truth that a machine is much more capable of performing this type of work accurately and quickly.

Upskilling: AI can support the upskilling and professional development of employees. Businesses can leverage AI-powered learning platforms to deliver personalized training, adaptive learning experiences and skill assessments. This allows businesses to foster a culture of continuous learning and adaptability, ensuring that employees possess the necessary skills for evolving job roles.

Changing Job Roles: AI will reshape job roles and create new opportunities, no doubt. While automation may eliminate certain repetitive tasks, it also leads to the emergence of new roles that require human-AI collaboration. Businesses can focus on leveraging human skills such as creativity, critical thinking, problem-solving, and empathy, which are complementary to AI capabilities. This encourages the evolution of job roles towards more strategic and value-added activities.

In conclusion, AI in business has become increasingly prevalent and transformative across various industries. AI technologies are revolutionising how organizations operate, make decisions, and engage with customers. From automation and process optimization to data analysis and customer personalization, AI is proving to be a powerful tool that drives efficiency, productivity, and innovation. While the integration of AI in knowledge businesses brings remarkable advantages, it also raises important considerations. Ethical concerns surrounding data privacy, algorithm bias, and transparency must be addressed to ensure the responsible and trustworthy use of AI in marine businesses. Striking a balance between the power of AI and the human touch in knowledge businesses is crucial to maintain the integrity and value of human expertise as the future unfolds.

Jeff Wilson has spent 38 years in the marine industry. Starting his career as a deck cadet with Shell in the 1980s, he is a Master Mariner with an MSc in Strategy and Economics who has held senior roles in ship owning and ship management and in the last decade occupied senior roles in London based marine consulting and survey companies.

After five years as Managing Director of a worldwide survey services Company, in October 2022 he moved to Van Ameyde Marine as Managing Director of Van Ameyde’s UK marine business.

Van Ameyde Marine is a global marine surveying and consultancy firm that offers marine services including cargo survey, casualty investigation, project cargo & heavy lift and loss prevention. With over 100 staff marine surveyors and consultants worldwide, Van Ameyde Marine is one of the largest single service providers for the marine industry.


Instagram Posts from the IIMS @iimsmarine